Software Engineer, Robotics Simulation & AI Infrastructure

Qualcomm•San Diego, CA
•$109,400 - $164,200

About The Position

Qualcomm Technologies’ Advanced Robotics Team is building an AI-first stack for general-purpose robots, combining heterogeneous compute, a Robotics SDK, an integrated simulation platform, and AI operations infrastructure. The team’s in-house simulator is built on high-performance computing and open data formats such as OpenUSD, URDF, and glTF, and is intended for Qualcomm robotics SoCs and diverse robot applications, including manipulation, locomotion, navigation, vision, and multi-stage tasks. This is primarily a simulation engineering role, with approximately 80% of the work focused on the simulator and 20% on enabling AI infrastructure workflows. The engineer will own significant parts of the platform, such as runtime architecture and abstractions, physics and rendering integrations, training and data-generation pathways, hardware-in-the-loop and CI systems, or developer tooling. The simulator supports interactive workstation authoring and vectorized GPU environments as well as orchestrated headless cloud jobs. It is used for policy training, synthetic data generation, release validation, hardware-in-the-loop testing, and reducing the simulation-to-reality gap. The position involves collaborating with AI operations, perception, controls, and silicon teams. The AI operations team owns the surrounding training, dataset, and deployment infrastructure; this role integrates simulation with those workflows rather than taking ownership of that infrastructure. The position includes producing design documents, reviews, tests, and APIs used by other teams.

Requirements

  • Bachelor’s degree or equivalent in Engineering, Information Systems, Computer Science, or a related field.
  • Strong software engineering fundamentals and production code that the candidate can discuss in depth.
  • Working grasp of robotics simulation concepts such as rigid-body dynamics, kinematics, coordinate frames, numerical integration, or sensor models, or the aptitude to develop that knowledge quickly.
  • Ability to learn unfamiliar technology stacks quickly and independently.
  • Effective use of modern AI tools in the software development workflow.
  • Comfort working in areas where some designs are established and others are still being validated.
  • Motivation to work on physical AI and robotics systems that function in the real world.

Nice To Haves

  • Experience with robotics simulators or real-time 3D engines such as MuJoCo/MJX, Isaac Sim or Isaac Lab, Newton, PhysX, Gazebo, Drake, Unreal Engine, or Unity; relevant experience in games, VFX, or autonomous-vehicle simulation is also applicable.
  • GPU programming and performance-engineering experience with CUDA, Warp, Vulkan, compute shaders, or accelerator-aware data layouts.
  • Machine-learning experience for robotics, including large-scale reinforcement learning, imitation learning, robot foundation models, vision-language-action models, or PyTorch.
  • 3D graphics and scene-pipeline experience, including OpenUSD composition, physically based rendering materials, real-time or offline rendering, or sensor simulation.
  • Robotics and distributed-systems experience, including ROS 2/DDS, URDF/MJCF, publish-subscribe systems, schema-driven wire formats, containers, cluster orchestration, or CI/CD for compute-heavy workloads.
  • Working fluency with AI infrastructure used alongside simulation, such as Kubernetes-native orchestration (Argo Workflows or similar), containerized GPU jobs, shared-storage datalakes using NFS/PVC volumes, config-driven experiment frameworks, checkpoint/resume semantics, or TensorBoard and Weights & Biases.
  • Effective use of modern AI tools, including frontier AI models, AI coding and agentic workflows, AI-assisted learning, and knowledge-base creation.

Responsibilities

  • Design and build core simulator subsystems, including scene representation and authoring, physics backends, sensor models, and rendering for robotics scenarios.
  • Profile and optimize simulation and training workloads, including physics solvers, rendering, and data pipelines, to support interactive use and high-throughput training.
  • Integrate and extend open engines for GPU physics, USD/Hydra rendering, and offline path tracing behind clean, swappable interfaces.
  • Build high-throughput policy-training and synthetic-data-generation pathways using vectorized environments, GPU-resident state, procedural scene variation, domain randomization, and ground-truth labeling.
  • Make simulation a software-release gate by building scenario suites, deterministic replay, benchmarks, and metrics to catch regressions before hardware deployment.
  • Integrate simulation as a configurable stage in collect–merge–train–eval workflows with reproducible run manifests and artifacts, collaborating with the AI operations team.
  • Support execution of the same experiments locally and as containerized GPU jobs using Argo Workflows on Kubernetes, without owning the surrounding infrastructure.
  • Establish hardware-in-the-loop configurations in which production robotics software runs on Qualcomm silicon and communicates with the simulator on a host workstation or in the cloud.
  • Measure simulation-to-reality divergence using system identification, contact and actuator modeling, sensor noise analysis, and transfer measurements.
  • Partner with AI operations, perception, controls, and silicon teams; own design documents, reviews, tests, and APIs that other teams depend on.

Benefits

  • Competitive annual discretionary bonus program.
  • Opportunity for annual RSU grants (employees on sales-incentive plans are not eligible for the annual bonus).
  • Competitive benefits package designed to support employees at work, at home, and at play.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service